1 / 20

Image Denoising via Learned Dictionaries and Sparse Representations

*. * Joint work with Michal Aharon . Image Denoising via Learned Dictionaries and Sparse Representations. Michael Elad The Computer Science Department The Technion – Israel Institute of technology Haifa 32000, Israel

hesper
Télécharger la présentation

Image Denoising via Learned Dictionaries and Sparse Representations

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. * * Joint work with Michal Aharon Image Denoising via Learned Dictionaries and Sparse Representations Michael Elad The Computer Science Department The Technion – Israel Institute of technology Haifa 32000, Israel IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR) New-York, June 18-22, 2006

  2. ? Remove Additive Noise Noise Removal ? Our story focuses on image denoising … • Important: (i) Practical application; (ii) A convenient platform (being the simplest inverse problem) for testing basic ideas in image processing. • Many Considered Directions: Partial differential equations, Statistical estimators, Adaptive filters, Inverse problems & regularization, Example-based techniques, Sparse representations, … Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  3. Part I: Sparse and Redundant Representations? Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  4. Relation to measurements Prior or regularization ThomasBayes 1702 - 1761 Denoising By Energy Minimization Many of the proposed denoising algorithms are related to the minimization of an energy function of the form y : Given measurements x : Unknown to be recovered • This is in-fact a Bayesian point of view, adopting the Maximum-Aposteriori Probability (MAP) estimation. • Clearly, the wisdom in such an approach is within the choice of the prior – modeling the images of interest. Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  5. Energy Smoothness Adapt+ Smooth Robust Statistics • Mumford & Shah formulation, • Compression algorithms as priors, • … Total-Variation Wavelet Sparsity Sparse & Redundant The Evolution Of Pr(x) During the past several decades we have made all sort of guesses about the prior Pr(x) for images: Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  6. Every column in D (dictionary) is a prototype signal (Atom). N • The vector is generated randomly with few (say L) non-zeros at random locations and with random values. N A sparse & random vector K A fixed Dictionary The Sparseland Model for Images M Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  7. D-y= - Our MAP Energy Function • We Lo norm is effectively counting the number of non-zeros in . • The vector  is the representation (sparse/redundant). • The above is solved (approximated!) using a greedy algorithm - the Matching Pursuit [Mallat & Zhang (`93)]. • In the past 5-10 years there has been a major progress in the field of sparse & redundant representations, and its uses. Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  8. D should be chosen such that it sparsifies the representations The approach we will take for building D is training it, based on Learning from Image Examples One approach to choose D is from a known set of transforms (Steerable wavelet, Curvelet, Contourlets, Bandlets, …) What Should D Be? Our Assumption: Good-behaved Images have a sparse representation Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  9. Part II: Dictionary Learning: The K-SVD Algorithm Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  10. D X  A Each example has a sparse representation with no more than L atoms Each example is a linear combination of atoms from D Measure of Quality for D Field & Olshausen (‘96) Engan et. al. (‘99) Lewicki & Sejnowski (‘00) Cotter et. al. (‘03) Gribonval et. al. (‘04) Aharon, Elad, & Bruckstein (‘04) Aharon, Elad, & Bruckstein (‘05) Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  11. D Initialize D Sparse Coding Nearest Neighbor XT Dictionary Update Column-by-Column by Mean computation over the relevant examples K–Means For Clustering Clustering: An extreme sparse representation Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  12. D Initialize D Sparse Coding Use Matching Pursuit XT Dictionary Update Column-by-Column by SVD computation over the relevant examples The K–SVD Algorithm – General Aharon, Elad, & Bruckstein (`04) Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  13. Part III: Combining It All Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  14. k D N Extracts a patch in the ij location Our prior From Local to Global Treatment • The K-SVD algorithm is reasonable for low-dimension signals (N in the range 10-400). As N grows, the complexity and the memory requirements of the K-SVD become prohibitive. • So, how should large images be handled? • The solution: Force shift-invariant sparsity - on each patch of size N-by-N (N=8) in the image, including overlaps [Roth & Black (`05)]. Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  15. What Data to Train On? Option 1: • Use a database of images, • We tried that, and it works fine (~0.5-1dB below the state-of-the-art). Option 2: • Use the corrupted image itself !! • Simply sweep through all patches of size N-by-N (overlapping blocks), • Image of size 10002 pixels ~106 examples to use – more than enough. • This works much better! Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  16. ComputeD to minimize using SVD, updating one column at a time Compute x by which is a simple averaging of shifted patches Computeij per patch using the matching pursuit Block-Coordinate-Relaxation x=y and D known x and ij known D and ij known Complexity of this algorithm: O(N2×L×Iterations) per pixel. For N=8, L=1, and 10 iterations, we need 640 operations per pixel. Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  17. Source Result 30.829dB Noisy image The obtained dictionary after 10 iterations Initial dictionary (overcomplete DCT) 64×256 Denoising Results The results of this algorithm compete favorably with the state-of-the-art (e.g., GSM+steerable wavelets [Portilla, Strela, Wainwright, & Simoncelli (‘03)] - giving ~0.5-1dB better results) Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  18. A Bayesian (MAP) point of view An energy minimization method Getting a relatively simple and highly effective denoising algorithm Using an image prior based on sparsity and redundancy Using examples from the noisy image itself to learn the image prior Today We Have Seen … More on this in http://www.cs.technion.ac.il/~elad Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  19. Fixing all A and D apart from the kth column, and seek both dk and the kth column in A to better fit the residual! We should solve: K–SVD: Dictionary Update Stage We refer only to the examples that use the column dk D SVD Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

  20. D D is known! For the jth item we solve XT K–SVD: Sparse Coding Stage Solved by Matching Pursuit Image Denoising Via Learned Dictionaries and Sparse representations By: Michael Elad

More Related